In recent years, the group researches are focusing on frameworks for integrating and coordinating heterogeneous software components and humans in complex distributed systems.

The systematic design, development and testing of such frameworks represent a vast and exciting area for research. Indeed, they must seamlessly integrate software components such as legacy applications; web services respecting various standards; as well as rule engines, software agents and even humans, who intervene to discover basic services and compose them in proper workflows. Furthermore, the knowledge necessary for these orchestration tasks is usually both procedural (i.e. composed of methods and functions) and declarative (i.e. composed of ontologies, rules and facts).

In this context, our group is currently working on several research projects such as:

• the MAIL (Multi-Agent Integration Layer) framework, which offers a middleware platform with semi-finished intelligent software personal assistants incorporating procedural and declarative knowledge, as well as various integration and coordination agents for improving the communication between the software infrastructure and human end-users in complex distributed information systems (e.g. the MediMAS prototype in the e-health domain);
• in the  Internet of  Things (IoT) world, specifying and building, in collaboration wih the ETH Zürich's Distributed Systems Group, of a novel RESTful version of an EPCIS (Electronic Product Code Information Server), that would enable RFID-based mashups. Such a solution would for instance enable people to browse RFID observations as they would browse web pages. Furthermore, it would enable client to query the EPCIS and subscribe to EPC events through simple Web requests on URLs and get the notifications in RSS feeds;
• the development of a coherent set of generic software tools coupled with an appropriate declarative representation for producing new "composite services" in a given domain (e.g. automatic management of an household) by first discovering adequate basic web services in the field of interest, and then orchestrating them and their human users' actions with a workflow engine;
• the study of the consequences of the new cloud computing paradigm on software architectures in general, with a special emphasis on proposing generic solutions for tackling their security issues.